It’s hard to argue that blockchain technology does not use a significant amount of power. It does. However, “a lot” is a relative term, and the amount of fanfare over blockchain’s power consumption is misplaced.
Admittedly, a lot of this information has already been stated by others, but these arguments cannot be restated enough, and the way in which this information is provided here is fairly unique. I also have my own criticisms of blockchain technology and cryptoassets. But I do think that these technologies have a lot of potential and they don’t benefit from ignorant criticisms that aren’t in any way justified.
One such criticism is the environmental concern about energy consumption. If one were to compare the nature of blockchain to the nature of payment processing services, and compare the energy consumption of blockchain, to the overall energy consumption of humanity, it becomes fairly obvious that blockchain is not the ecological villain that some make it out to be.
More Than Just Transactions
One of the arguments that people use to criticize the energy consumption of blockchain is by comparing it to payment processing costs by institutions like VISA. However, blockchain is not simply a payment processing system. It is a distributed and decentralized ledger. It stores the entire transaction history of its users, in an easily accessible way. It therefore acts not only as payment processor, but transaction database and interface for accessing your transaction details.
Moreover, Ethereum and other blockchain technologies implement digital smart contracts which manage a lot of activities, largely relating to the creation and control of digital assets. Therefore it is far from reasonable to equate blockchain technology with payment processing and compare the energy consumption of the two. It would instead require comparing blockchain to the entire energy consumption of the financial industry, including the cost of running banks, ATMs, the backend servers that operate banking websites and all banking transactions, the distribution physical money, the armored cars that do so, the training of officers that secure them, and so on.
Total Blockchain Energy Consumption
It’s hard to say exactly how much energy the entirety of blockchain technology consumes. But I’m going to make a few simplifying assumptions and make a rough guess.
According to Vox, Bitcoin alone consumes about 67 terawatt-hours per year. This figure seems quite high, and it is. Proof of work architecture is energy intensive, and it’s hard to prevent it from being otherwise, because it needs computationally intensive problems to keep the system very secure. Ethereum, which has a very different use case, and different way of operating, burns another 7–21 TWh per year, according to Digiconomist. Let’s go with 21 TWh, so figure a high side estimate. Now, these two blockchains are some of the biggest. Let’s say all in all, the total power consumption of blockchain is ten times the total of those two, or 880 TWh per year. To add a further buffer, let’s round that to 1PWh (1,000 TWh).
This figure is an incredible amount of power. In the United States, residential power consumption is roughly 11,000 KWh per year, based on EIA figures. In other words, the energy used for blockchain technology could run 80 million households. But using household figures is silly. Most energy consumption is commercial and industrial, not residential.
Global Energy Consumption
Now that it’s clear just how much energy blockchain consumes, let’s compare it to how much people consume overall. If blockchain consumption is a large component of total energy consumption, then it’s reasonable to call it out as an energy hog, but if not, then such accusations are uncalled for, especially considering that blockchain technology does not simply act as a payment processor, but as a distributed and decentralized ledger that stores the history of all transactions, so that they can be accessed at a future date, and in many cases a robust contract management system for the handling of digital assets.
Enerdata has a lot of information on global energy consumption. It would be nice if it had a quick access figure for total energy consumption in TWh, but it does have breakdown by region, in MTOE (mellions of tonnes of oil equivalent). In 2018, the total was 13,978 MTOE. In total, that amounts roughly 162 PWh of energy consumption in 2018. So now there’s something to compare the initially phenomenal looking energy consumption of blockchain. The estimated total energy consumption of blockchain amounts to 0.6% of the world’s total energy consumption. Well, that certainly makes blockchain look a bit less like an energy hog, doesn’t it? Moreover, the average annual growth rate of global energy consumption, between 2008 (when Bitcoin first came to be) and 2018, was about 1.7% which is almost three times the total annual energy consumption of blockchain.
While it would be great to drive down the energy consumption of blockchain technology, and provide the technology to more users across the world, the accusation that blockchain technology is such an energy hug is uncalled for. Not only is the entire energy consumption of blockchain technology only about 0.6% of global energy consumption, but global energy consumption grows by about three times the total blockchain consumption rate, every single year. Moreover, blockchain technology is not simply a method through which economic transactions occur. It’s an entire system that manages the transaction history, as well as robust digital contracts that can handle long term digital asset management, and access to transaction history.
I honestly don’t know why this kind of misinformation is being promoted. I’m not sure if it’s some intentional tactic by the financial industry to attack blockchain. It could be. There are a lot of people who are threatened by blockchain and cryptoassets. However, Hanlon’s razor suggests that one should never attribute to malice that which is adequately explained by stupidity. Perhaps someone just saw the raw numbers and thought that they were more significant than they actually are.